AI.Reverie: Synthetic Data for Smarter AI

We live in a digital age where Artificial Intelligence is beginning to play a significant role. AI is considered to be a boon by some and bane by many. Misinformation has shrouded what it is truly capable of. Many believe that AI can really make the world sustainable. NY-based AI.Reverie thinks that this is possible through its synthetic data approach, which harnesses computer-generated, three-dimensional environments to accelerate machine learning in a controlled, responsible, and empirical manner.

Founded in 2017, AI. Reverie wants to change the perception of AI. Its synthetic data solutions perhaps provide a way to do this. The company’s solutions provide a virtually limitless supply of scenarios that are proven with benchmarked evidence and the measurable success of its business partners.

AI.Reverie was created with certain questions in mind: “What if we could predict – and then solve – the number-one obstacle facing AI and machine learning?” For this to happen, it was crucial to accelerate machine learning, the source of that learning – data – to be readily available. On one hand, the modern world generates seemingly infinite data. On the other, existing data fails to capture many real-world scenarios, and labeling that data by hand is labor-intensive and expensive. AI.Reverie set out to build a source of data with endless variety and a labeling capability that would resolve issues of time and expense.

Today, the company’s synthetic data learning environments and products has changed the paradigm of machine learning, creating a virtually infinite raw material (data) that has completely changed the cost structure of investment in machine learning for our clients and partners. AI.Reverie presents a collaborative, iterative approach grounded in synthetic data tools. It also provides models trained on its synthetic data that have begun to outperform those trained on real data alone.

The company recently raised $5.6 million in an equity round of financing. Vulcan Capital led the round, and other institutional investors included Compound, In-Q-Tel, Resolute Ventures, SGInnovate, TechNexus and Triphammer Ventures. The investment brings AI.Reverie’s financing since launch to over $10 million. YB Choi of Vulcan Capital joined AI.Reverie’s Board of Directors as part of this round.

AI.Reverie has had a brilliant 2019 and 2020 as it rapidly grew its client portfolio that ranged from government agencies to Fortune 500 clients. Their platform propelled projects in Defense, Retail, Smart Cities, Industrials, and Agriculture, including airport simulation, weapons detection, cashier-less shopping, and delivery bots. “We have now seen our technology solve the real challenges of a broad client base,” said Daeil Kim, co-founder and CEO. “Our platform is fine-tuned and ready to grow.” In October 2020, AI.Reverie was named as a Gartner Cool Vendor for 2020 within the area of AI Core Technologies.

Working with AFWERX

AFWERX, the U.S. Air Force's innovation arm, which was established in 2017, has been working overtime to scale the innovations for the country’s defense forces. As part of this, the organization has rewarded Ai.Reverie with contracts.

Earlier this year, AI.Reverie was awarded AFWERX Small Business Innovation Research (SBIR) Phase I contract to enhance the Department of Defense's computer vision models. The contract was awarded to the NY-company as it was looking for solutions that would accelerate reconnaissance to the speed required in a contingency environment. Currently, the computer vision models that power intelligence-gathering must be trained on data gathered from classified locations and hard-to-reach places. Pentagon AI experts have detailed the high-cost, labor intensive labeling process the data must undergo before it can be put to work. AI.Reverie’s synthetic data platform resolves the challenges of rare data and expense by automatically generating millions of fully annotated, richly diverse images — quickly and at a low cost.

AI.Reverie was also awarded a $1.5 million Phase 2 Small Business Innovation Research (SBIR) contract by AFWERX to build AI algorithms and improve navigation capabilities for the United States Air Force. “We are delighted to be awarded this Phase 2 SBIR contract and look forward to partnering with the U.S. Air Force’s 7th Bomb Wing on this project,” said Daeil Kim, CEO and Co-founder of AI.Reverie. “Our goal is to equip soldiers to successfully navigate any situation, and we are proud to contribute to the United States military’s mission to protect our country.”

The AI Expert

Daeil Kim, Founder and CEO

A New Yorker, Daeil is the technical founder of AI.Reverie. Prior to AI. Reverie, he worked with the New York Timesas Senior Data Scientist.He has been a Research Associate at Mind Research Network too.

He holds a Ph.D in computer science from Brown University, and his research has been published in several of the top machine learning conferences (NIPS, ICML, AISTATS).

“AI.Reverie generates an entirely new class of data at scale that makes AI training affordable, fast and productive.”

“We’re journalists, computer scientists, and creative thinkers. We bring different approaches to a shared vision in which artificial intelligence solves for both business and human needs.”

“We’re believers in the future of AI through synthetic data, and we’ve made it real.”